control board: reply from the board through agent-send.sh (#1505)
Piece 2 of the MVP (#1503). A one-line reply box and Send in the detail of rows with a live registration; POST /api/reply runs tools/tmux/agent-send.sh -s <session> -S <host>:control-board [-L <socket>] -m <text> once for one seat and returns the exit code, stdout and stderr. The page shows delivered or failed with the tool's stderr; other rows say "reply needs a registered seat". No send-keys, queue, retries, history or broadcast; packages/seat and agent-send.sh untouched. Every message ends with a fixed trailer telling the seat to answer in its own session (Jason's refinement after the first Gate C exchange; the board has no pane). Board suite 98/98. Co-Authored-By: Claude Fable 5.1 <[email protected]>
This commit is contained in:
Claude Fable 5.1
@@ -223,3 +223,4 @@ are never rewritten or removed; corrections are new entries.
|
||||
- 2026-09-12 — darkwing (Claude Code, coordinator; formerly jarvis) — Seat registration (#1504), Jason's go on the professor session's brief: new `packages/seat` and `scripts/mosaic` (`launch <seat>` registers then execs the launch script unchanged; `seat task` edits the task); board reads `<dataRoot>/seats/<layout>/<seat>/registration.json` and lets registered task/project/workspace override derived values with a source tag; the four repo launch scripts register themselves; fleet one-liner written for orch-01. Tests: seat 15, board 79 → 89, launch scripts 5, registry 69, config 24. Sonnet review found a seat-name collision across layouts; fixed by keying on layout, second pass APPROVED. Live check on a scratch server against a temp data root. Also found Jason's launch refused on the Pi pin (node_modules 0.84.4 vs pin 0.85.1 since 557aba0f); `npm ci` fixed it, the refusal now names both versions; the pin stays unless Jason decides otherwise.
|
||||
- 2026-09-12 — darkwing (Claude Code, coordinator) — #1504 follow-up on two defects the professor session reported and this session verified: the darkwing launch test leaked fixture registrations into the real data root (flock spawn without the fixture config; fixed and guarded so a leak fails), and a refused launch left a dead-pid record the board honoured (scanner now probes the pid; stale records are shown but do not override). Fleet launchers marked not planned per Jason. Board 90/90, seat 15/15, launch scripts 5/5. Committed and pushed to refactor.
|
||||
- 2026-09-12 — darkwing (Claude Code, coordinator) — Control board: model per row (#1503) on Jason's ask: readSession tracks model/provider from the latest model_change entry or assistant turn; page shows the model under the agent name and a Model detail row. Live check on a scratch board: all 42 real rows named a model. Board 91/91. Committed and pushed to refactor.
|
||||
- 2026-09-12 — darkwing (Claude Code, coordinator) — Control board piece 2, reply-from-board (#1505, brief by Jason): `POST /api/reply` runs `tools/tmux/agent-send.sh` for one registered, live seat and returns the exit code and streams; the page has a reply box in registered live rows' detail with delivered/failed receipts. Live check on a scratch tmux session through the real agent-send.sh: delivered, then a failure receipt with the tool's stderr. Board 98/98. Gate C first exchange ("pizza?") delivered and read back within one scan; the seat's send-back to control-board was refused as expected, so every board message now ends with a fixed trailer (Jason's ruling via the professor session). Committed and pushed to refactor.
|
||||
|
||||
@@ -407,3 +407,67 @@ of `gpt-5.6-*` and `glm-5.3`). Board suite 91/91. Jason's board on 7331
|
||||
needs a restart to pick this up. Sonnet review found the model hover title
|
||||
escaped an already escaped string (double encoding for names with `&` or
|
||||
`<`); fixed before commit, then APPROVED on that basis.
|
||||
|
||||
**2026-09-12 — Piece 2: reply-from-board (#1505).** Built to the brief
|
||||
above. `POST /api/reply {agent, text}` rescans, finds the row, and runs
|
||||
`tools/tmux/agent-send.sh -s <session> -S <host>:control-board -m <text>`,
|
||||
plus `-L <socket>` when the registration names one, with
|
||||
`MOSAIC_TMUX_SOCKET` stripped from the tool's environment so the
|
||||
registration is the only source of the socket. The answer carries
|
||||
`delivered`, the exit code, stdout and stderr; a non-zero exit is
|
||||
reported as is. Refusals before the tool runs: bad or empty or long text
|
||||
400, unknown row 404, no registration or stale registration or no tmux
|
||||
session 409, tool not startable 500. The page puts a one-line box and
|
||||
Send in the detail of rows whose registration is live and has a tmux
|
||||
session; other rows say "reply needs a registered seat". The receipt
|
||||
reads `delivered <time> to tmux <session>` or `failed (exit N): <stderr>`;
|
||||
drafts and receipts survive the periodic refresh and the caret is given
|
||||
back to the box; a delivered reply clears the box and rescans at once.
|
||||
Not built, per the brief: raw send-keys, queue, retries, history,
|
||||
broadcast, more than one seat per request; no change to `packages/seat`,
|
||||
the registration record or `agent-send.sh`.
|
||||
|
||||
Live check on a scratch server (port 7332) with a scratch tmux session
|
||||
`cb-reply-test` running `cat` and a temp data root: `agent-send.sh`
|
||||
accepted `dragon-lin:control-board` as sender, the text arrived in the
|
||||
pane as `[dragon-lin:control-board -> dragon-lin:cb-reply-test] ...`, and
|
||||
the page showed `delivered`. Killing the session also killed the
|
||||
registered pid, so the row went stale and the box gave way to "reply
|
||||
needs a registered seat", as intended. With a live pid pointed at a
|
||||
missing session the receipt read `failed (exit 1): ERROR: invalid
|
||||
resolved pane identity` and the draft stayed in the box. Tests: a fake
|
||||
`agent-send.sh` on the server's configured path captures argv and the
|
||||
environment; cases for `-s`/`-S`/`-m` without `-L`, with `-L`, exit 4 with
|
||||
stderr, each refusal with the tool never called, a missing tool, and
|
||||
static checks on the page (box only when `alive !== false` and a tmux
|
||||
session, JSON POST, escaped receipts). Board suite 98/98 (was 91). Gate C
|
||||
is Jason's to run against a real waiting seat; his board on 7331 needs a
|
||||
restart first.
|
||||
|
||||
Sonnet review of the piece: one finding, the route comment at the top of
|
||||
`serve.mjs` had the `/api/reply` line spliced into the middle of the
|
||||
`/api/seen` description, so it read as if a reply rescanned and returned
|
||||
the index; the comment was split back into two entries before commit.
|
||||
Noted but not changed: the body limit is bytes and the text limit is
|
||||
characters, so a non-ASCII text near 2000 characters is refused as an
|
||||
oversized body rather than as long text; and a row open in both "Waiting
|
||||
on you" and its project group has two forms with the same id, so the
|
||||
refocus after a refresh may land on the other copy.
|
||||
|
||||
**2026-09-12 — Gate C, first exchange, and the trailer.** Jason answered
|
||||
darkwing from his restarted board with "pizza?". The delivery worked
|
||||
(`[dragon-lin:control-board -> dragon-lin:darkwing] pizza?` in darkwing's
|
||||
session log) and the row showed the answer as its last message within one
|
||||
scan, so the return path exists: the board reads the transcript. What
|
||||
failed was the seat's side: the communications skill told the seat a reply
|
||||
only exists once `agent-send.sh` runs, the seat aimed at `-s
|
||||
control-board`, and the tool refused because `control-board` is not a
|
||||
tmux session ("invalid resolved pane identity"). That refusal is correct;
|
||||
the board is not made a pane. Fix in the board, relayed as Jason's ruling
|
||||
by the professor session: every message the board sends ends with a fixed
|
||||
line, `(control-board: answer in your own session as usual; the board
|
||||
reads your transcript. Do not agent-send to control-board.)`, exported as
|
||||
`REPLY_TRAILER`; `-S` stays `<host>:control-board` so the preamble stays
|
||||
honest about the origin. The fake agent-send test now checks the message
|
||||
text ends with the trailer; the receipt is unchanged. Committed with the
|
||||
piece.
|
||||
|
||||
+21
-7
@@ -32,13 +32,27 @@ Package: `packages/seat`. Fleet seats stay on their own launchers (Jason's
|
||||
ruling, 2026-09-12); only `agents/` seats register. A record whose pid is
|
||||
gone is stale and does not override the derived values.
|
||||
|
||||
Next action: step 3 of #1503 continues — daily use with #1504 landed. Jason
|
||||
launches his seats through the new path (or the launch scripts, which now
|
||||
register), sets tasks with `seat task`, and reports what is wrong, missing
|
||||
or confusing; each report becomes a small fix with a test, committed and
|
||||
pushed to `refactor`. The sequence Jason set stays: seat registration (done),
|
||||
then reply-from-board, then the WebUI on Dewey's Console design absorbs the
|
||||
board as its first screen. Close #1503 when Jason says the page answers "who
|
||||
Gate B passed 2026-09-12 (seat registration shown and used on the board).
|
||||
|
||||
Piece 2, reply-from-board (#1505; brief in the plan page, section "Piece
|
||||
2: reply-from-board", approved by Jason 2026-09-12), is built and pushed
|
||||
2026-09-12: a text box and Send in the detail of registered, live rows;
|
||||
`POST /api/reply` runs `tools/tmux/agent-send.sh -s <session> -S
|
||||
"<host>:control-board" [-L <socket>] -m <text>` and returns the exit code,
|
||||
stdout and stderr; the page shows `delivered` or `failed` with the stderr.
|
||||
No send-keys, queue, history or broadcast; no change to `packages/seat` or
|
||||
`agent-send.sh`. Board suite 98/98.
|
||||
|
||||
Gate C's first exchange happened 2026-09-12 ("pizza?" to darkwing:
|
||||
delivered, read back within one scan; the seat's send-back to
|
||||
`control-board` was refused as expected, so every board message now ends
|
||||
with a fixed trailer telling the seat to answer in its own session).
|
||||
|
||||
Next action: Jason confirms Gate C on the trailer build (restart the board
|
||||
on 7331, answer a `waiting` seat, the row turns `working` with the reply
|
||||
as its newest user message and the seat answers in its own session).
|
||||
Record the pass or the failure in the plan page and here. Then the WebUI on Dewey's Console design absorbs
|
||||
the board as its first screen. Close #1503 when Jason says the page answers "who
|
||||
is waiting on me" without him opening a terminal. Out of scope until he
|
||||
asks: auth or provider registry, roster schema changes, hooks/plugins, comms,
|
||||
memory, multiple sessions per seat, stopping seats, new root files. The
|
||||
|
||||
Reference in New Issue
Block a user